{"id":"k2-fsa-omnivoice","nom":"OmniVoice","editeur":"Next-gen Kaldi","pays":null,"poids_ouverts":true,"poids_selon":null,"licence":"CC BY-NC","licence_selon":null,"liens":{"page":"https://huggingface.co/k2-fsa/OmniVoice","essayer":"https://huggingface.co/spaces/k2-fsa/OmniVoice"},"fiche_auto":true,"page":"https://quelleia.com/voix/modeles/k2-fsa-omnivoice/","releve":"2026-10-19","en_bref":"1er sur 16 en prononciation du français.","presentation":"Modèle de synthèse vocale multilingue en zéro-shot, qui clone une voix à partir d'un court extrait audio et peut aussi créer une voix à partir d'attributs (genre, âge, accent, chuchotement). L'équipe annonce 646 langues et une architecture de type modèle de langage à diffusion, initialisée avec les poids de Qwen3-0.6B.","sortie":"2026-04-01","langues":{"annoncees":{"annoncees":"646 langues, dont le français","nombre":646,"francais":true},"prises_en_charge_selon_open_tts":646,"evaluees":["fr"]},"preuves":[{"appuie":["presentation"],"lu":"https://huggingface.co/k2-fsa/OmniVoice/raw/main/README.md (« OmniVoice is a massively multilingual zero-shot text-to-speech (TTS) model supporting over 600 languages. Built on a novel diffusion language model-style architecture »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/k2-fsa/OmniVoice/raw/main/README.md (« **Voice Design**: Control voices via assigned speaker attributes (gender, age, pitch, dialect/accent, whisper, etc.). »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://arxiv.org/html/2604.00688v1 (« which is initialized with the pre-trained LLM weights of Qwen3-0.6B »), 17/10/2026"},{"appuie":["sortie"],"lu":"https://arxiv.org/html/2604.00688v1 (« arXiv:2604.00688v1 [cs.CL] 01 Apr 2026 OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Models »), 17/10/2026"},{"appuie":["langues"],"lu":"https://raw.githubusercontent.com/k2-fsa/OmniVoice/master/docs/languages.md (« OmniVoice supports **646 languages** with a total of **581k hours** of training data. »), 17/10/2026"},{"appuie":["langues"],"lu":"https://raw.githubusercontent.com/k2-fsa/OmniVoice/master/docs/languages.md (« | 180 | French | fr | fra | »), 17/10/2026"}],"classements":[{"id":"synthese/prononciation","nom":"Prononciation du français","rang":1,"sur":16,"valeur":4.72,"unite":"%","chiffres":{"rtfx":33.57,"taille_milliards":0.81,"langues":646,"clonage":true},"voisins":["fish-audio-s2-pro"],"source":{"nom":"Open TTS Leaderboard","mention":"Source : Open TTS Leaderboard, Hugging Face","url":"https://huggingface.co/spaces/hf-audio/open_tts_leaderboard","releve":"2026-10-19"}}],"noms_dans_les_sources":{"hf_audio":"k2-fsa/OmniVoice"}}